st: bootstrapping average treatnment effect uing cluster option

["Dawit Lidia" <ghibertmk@hotmail.com>]

Dear all,

I�m estimating the average treatment effect on the treated (ATT) using attk 
command . I have  multiple observations per group. Observations are not 
independent and I want to adjust intra-group correlation. The attk as 
developed by Ichino and Becker does not support cluster option. However, I 
tried the following and obtained some results.

bs"attk   outputvalha     con    solfert1 solfert2   sodepth2 sodepth3   
lnha meslope  homedist1 homedist3 rentdum   oxtluha  tottluha visit age  
age2  fwfha fwf2ha mwfha    landae      ,  comsup    bwidth(0.06)" r(attk) , 
reps(200) cluster(hhno)

The ATT from this command is the same as the ATT I obtained without cluster 
option but the standard errors differ.

My question is, is the above specification is correct?

Many thanks.
Dawit

_________________________________________________________________
MSN Messenger http://www.msn.no/computing/messenger - Den korteste veien 
mellom deg og dine venner

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/